Nina Isabelle Onia
Professor Doris Huber
ENGL 109W-03
2 November 2022
Rhetorical Analysis
In academic writing, rhetoric is the use of language to sway an audience to understand an author’s thesis—their main idea—presented in a text. An author establishes rhetoric through credibility, connection to the audience, concise discussion of their thesis, and reinforcement of their intent behind writing a specific article. This rhetorical analysis will cover how the article “Methods to Optimize Carbon Footprint of Buildings in Regenerative Architectural Design with the Use of Machine Learning, Convolutional Neural Network, and Parametric Design” by Mateusz Ploszaj-Mazurek, Elżbieta Ryńka, and Magdalena Grochulska-Salak manages to establish rhetoric. In this scholarly article, three authors of varying professional experience levels clearly discuss their thesis to a wide audience of readers with clear explanations of complex terms and analyses of graphics that accompany the text.
All three authors of this scholarly article have some professional relation with architecture, sustainability, and AI. Mateusz Ploszaj-Mazurek recently graduated from Politechnika Warszawska in 2021 with a Ph.D. in architecture and currently works at the architecture firm BJERG ARKITEKTUR. Out of all of the authors, he has the least documented professional writing. Including the article in this analysis, he has only written two scholarly articles. However, Ploszaj-Mazurek’s professional writing and architectural projects focus on machine learning—a form of AI—and optimizing designs to reduce carbon footprint. He is the only author in this trio who has the experience and knowledge to explain AI to a wide audience. By contrast, Elżbieta Ryńka has the most documented experience with sustainability. She is a senior consultant at Go4Energy: a Polish real estate company that aids other companies in switching to more energy-efficient design practices. Ryńka has been writing scholarly articles since 2008. Topics she discussed include biodiversity, urban planning, and developing a sustainable design. Ryńka has the most incentive to write this article because she has taken a stance toward more eco-friendly architectural projects. Lastly, Grochulska-Salak meets these authors in the middle in terms of experience. Similar to Ploszaj-Mazurek, she has a Ph.D. in architecture. Similar to Ryńka, she has written several scholarly articles about sustainability and, as such, is a viable author for this article. Topics unique to Grochulska-Salak include urban agriculture and blue and green infrastructure. All three authors are from Warsaw, Poland and are affiliated with the Warsaw University of Technology. There, Ploszaj-Mazurek is an assistant teacher and researcher, Ryńka is a professor and the Chair of Urban Design and Physical Planning, and Grochulska-Sala was an assistant professor for the Department of Civil Engineering, retiring in 2021. These three alumni have the scholarly credentials to teach new members of the discipline through academic writing.
In this article, the authors present information to an audience who are unfamiliar with sustainability, AI, and architecture but want to learn more. Their target audience consists of students who study in the same disciplines the authors enrolled in and taught. As such, Ploszaj-Mazurek and others begin the article by defining terms that are mentioned in the title. For instance, the audience may not know what “The Total Carbon Footprint” is. First, the authors define “Carbon Footprint” as “a measure of the exclusive amount of carbon dioxide emissions that is directly and indirectly…accumulated over the life stages of a product.” Then, they direct the audience to a graphic that displays the life stages of a building—a type of product—that are signified with a letter and number (e.g. A4 for “Transport” during the construction phase) (Ploszaj-Mazurek et al. 1-2). By establishing the professional, more advanced concepts first, the authors can neatly present the information once the audience understands what they are talking about. Another way the authors connect to their audience is by relating their presented research to people of the audience’s skill level. Later on in the introduction, Ploszaj-Mazurek and others claim that “architecture students…create better architectural designs in terms of global warming and other environmental impacts” if they use the right programs when designing” (2). The authors assume the audience does not know about their potential in the practice. Architecture students are new to the discipline and are likely unaware of the environmental issues that arise from it. The authors establish that the following information in the article is worth paying attention to because it will help their audience understand what different programs can benefit their architectural projects. When the authors delve into the studies, they continue their process of explaining a term or study as a quick pause between recounting the study. For example, after discussing the parameters of a model in the first study, Ploszaj-Mazurek and others direct the audience to a table. This table lists specific measurements and materials that the model is made of. The authors then discuss heating and ventilation, information that is not listed in the table (Ploszaj-Mazurek et al. 5). Here, now that the authors have reached the body of the article, they do not hold the readers’ hands as tightly. Supposedly, the audience should have some idea of the components of a building and the limitations in constructing it. With this in mind, the authors take the time to add context without going on too long of a tangent. This allows for the article to break down the discussion into neat segments.
This scholarly article is organized into sections and subsections that break down the meaning of the title—“Methods to Optimize Carbon Footprint of Buildings in Regenerative Architectural Design with the Use of Machine Learning, Convolutional Neural Network, and Parametric Design”—with text and graphics. The article comes with an abstract: an overview of the contents of the article. Ploszaj-Mazurek and others briefly mention major concepts later including Carbon Footprint, Machine Learning (ML), and Convolutional Neural Networks (CNN) (1). The article then begins with an introduction that is broken down into three subsections: “Regenerative Design,” “Architectural Design Process,” and “Artificial Intelligence in the Architectural Design Process.” Ploszaj-Mazurek and others put these topics in this order to start with the broad subject of regenerative design, segue to how architectural projects are planned, and end by applying regenerative design to architecture through AI (2-3). Explaining the title first provides plenty of context for the bulk of the case studies. The next section is “Materials and Methods.” Here Ploszaj-Mazurek and others provide a figure of a parametric model that the first study is based on, insinuating that the second and third studies will modify that model. They also discuss the specific AI used, such as “Grasshopper, a parametric modelling plugin” that made the model (Ploszaj-Mazurek et al. 4). This brief section provides context for the three studies in the “Results” section. The first case study has three subsections. Throughout these sections, the authors refer back to concepts mentioned in the introduction now with the context of the study, such as using AI to calculate the “Embodied Carbon (A1-A3 Lifecycle stages)” of the “1500 simulations” of the parametric model (Ploszaj-Mazurek et al. 6). The second study is broken up into two sections. The first section transitions from the ideal version of the first study model that the has the lowest possible Carbon Footprint according to the AI. The second section provides charts and figures that display how the new model opens opportunities for experimenting with the model’s shape and testing the accuracy of the AI’s calculations of the Total Carbon Footprint (Ploszaj-Mazurek et al. 6-9). The third and final study is made up of four sections. The first introduced the CNN AI that had yet to be applied to the model. The second recounted the AIs’ outputs of the model in randomly generated urban landscapes. The third is a discussion of the results and accuracy of the AI similar to the format of the second study. The fourth is a discussion applying the results of the study to real world applications of AI in architecture (Ploszaj-Mazurek et al. 11-15). The authors pair text with visual guides in a way that portrays the purpose of their article efficiently to readers.
The most vital component of the collective rhetoric of these three authors is the purpose for writing this article: carbon emissions can be reduced using AI. The authors delve right into the exigence in the introduction: “[t]he built environment is a major contributor to the greenhouse gas emissions problem…a major part of these emissions stems from currently-applied design practices” (Ploszaj-Mazurek et al. 1). By beginning the article with a direct relation to architecture and environmental issues, the authors clearly indicate the importance of this scholarly article. They later mention remedies to said issues, such as “to proceed from sustainable architecture towards a restorative approach that can positively impact the environment,” or to “raise awareness concerning…the urban ecosystem” (Ploszaj-Mazurek et al. 2-3). The authors reinforce their thesis with broader concepts that are not necessarily tied to professionals in their discipline. That being said, transitioning to the body involving the case studies, Ploszaj-Mazurek and others then discuss the importance of AI. The primary AI they use for their studies is ML. Although AI is established as a quick and easy-to-use tool for professional architects and designers, the authors preface that ML is “limited by one crucial element, namely access to data” (Ploszaj-Mazurek et al. 4). The authors have presented a crucial counterargument that they tie in to their message: AI is an imperfect tool that requires the right data and appropriate application to work properly. This is reinforced by the inclusion of a margin of error between what the AI predicted and what human workers calculated. Ploszaj-Mazurek and others note that the first study had a margin error of 1.3% (8). Moving on to the second study, they imply that “extending the [ML] algorithm towards non-cuboid shapes” has not been tested frequently, urging them to test Grasshopper’s limits (Ploszaj-Mazurek et al. 9). The authors note that, although AI seems to have endless possibilities as to what it can create, claiming that AI has this power is redundant if it is not tested and analyzed. Moving on to the final study, the authors provide another counterargument applicable to the previous studies: it is unrealistic to plan an optimally eco-friendly building when there is no data on the site—the area the project will be built—provided to the AI (Ploszaj-Mazurek et al. 11). The authors’ intent is not to provide a step-by-step tutorial on how to use AI to instantly create a building with a low carbon footprint. Their intent is to apply the concept of using AI to sustainable design with the parameters set by the first study, urging other designers and architects to further these experiments and test the efficiency of AI.
Providing a strong rhetoric is not only a means of clearly stating and reiterating a these. An author should be self-aware and willingly present counterarguments and preface that they may not have all the data to back up their claim. The three authors of this scholarly article do just that. Each author has their own resume of professional work, some more extensive and diverse than others. The authors stick with their strengths and focus on the information they could gather. These imperfections allow the authors to connect to the audience, both parties denying that they are know-it-alls in the discipline and may need refreshers on acronyms and practices. This scholarly article encapsulates the authors’ rhetoric by displaying a level of knowledge and understanding of ML and sustainable architecture that allows readers to comprehend the thesis and consider their own stance on the matter should they research any of the topics further.